Fix naming of segment_id / utterance_id / recording_id to be more Kaldi-standard

CoEDL / elpis

🙊 software for creating speech recognition models.

https://elpis.readthedocs.io/en/latest/

Apache License 2.0

152 stars 33 forks source link

Open benfoley opened 2 years ago

benfoley commented 2 years ago

Seems that utterance_id is used in elpis/engines/common/output/ctm_to_elan.py whereas it should be recording_id.

Segment_id is being used for what is actually the utterance id.

Segment file is made of <utterance-id> <recording-id> <segment-begin> <segment-end>

Also, check the ctm_to_textgrid file.

alexis-michaud commented 2 years ago

This Issue could safely be closed, couldn't it? Now that HF is used & not Kaldi any longer. :leaves: :fallen_leaf: